• IEMEK Journal of Embedded Systems and Applications
• /
• v.19 no.3
• /
• pp.151-158
• /
• 2024

Although on-device artificial intelligence (AI) has gained attention to diagnosing machine faults in real time, most previous studies did not consider the model retraining and redeployment processes that must be performed in real-world industrial environments. Our study addresses this challenge by proposing an on-device AI-based real-time machine fault diagnosis system that utilizes continual learning. Our proposed system includes a lightweight convolutional neural network (CNN) model, a continual learning algorithm, and a real-time monitoring service. First, we developed a lightweight 1D CNN model to reduce the cost of model deployment and enable real-time inference on the target edge device with limited computing resources. We then compared the performance of five continual learning algorithms with three public bearing fault datasets and selected the most effective algorithm for our system. Finally, we implemented a real-time monitoring service using an open-source data visualization framework. In the performance comparison results between continual learning algorithms, we found that the replay-based algorithms outperformed the regularization-based algorithms, and the experience replay (ER) algorithm had the best diagnostic accuracy. We further tuned the number and length of data samples used for a memory buffer of the ER algorithm to maximize its performance. We confirmed that the performance of the ER algorithm becomes higher when a longer data length is used. Consequently, the proposed system showed an accuracy of 98.7%, while only 16.5% of the previous data was stored in memory buffer. Our lightweight CNN model was also able to diagnose a fault type of one data sample within 3.76 ms on the Raspberry Pi 4B device.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.7 no.6
• /
• pp.301-309
• /
• 2012

In this paper, a real-time monitoring system based on WSN is designed and implemented to monitor livestock growth environment information which includes the temperature, humidity and harmful gases such as $CO_{2},\;CO,\;NH_{3},\;H_{2}S$ and so on. The proposed system consists of the wireless sensor nodes, the monitoring management device, the management server and the user interface program based on PC/Smart phone. To verify the performance of the implemented system, gas measurement experiments are performed in laboratory environment by using the designed wireless sensor nodes. And it is able to estimate the concentration of gases. The implemented system is able to monitor the proposed environmental element information through the developed GUI.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.60 no.2
• /
• pp.99-104
• /
• 2011

In this paper, the intelligent landscape lighting power control and monitoring system with solar sell generator equipment is proposed. The first, the intelligent landscape lighting power controller is designed using the fuzzy logic control method. And the fuzzy logic controller is used to save power consumption for various reference intensity of the illumination. The second, the GUI monitoring system is presented. It has control and display faculty. And the practical experiment device is used to evaluate the performance criteria of the proposed intelligent landscape lighting power control system with the solar cell power generation equipment. From the experiment results, we present the property of proposed fuzzy controller such as steady state error, the tracking and power consumption characteristic for the reference intensity of illumination. And also we show the superiority of power control as well as the characteristic of GUI monitoring system in the proposed system.

• Proceedings of the IEEK Conference
• /
• 1999.11a
• /
• pp.925-928
• /
• 1999

A multi-channel gas leakage monitoring system was designed and fabricated by using CPLD(complex Programmable Logic .Device) for monitoring and controlling the leakage of natural gas from supplying-pipes under the ground. Fabricated SnO$_2$thick film gas sensor elements were attached on safeguard steel plate of natural gas supplying pipes, and the local monitoring system in this study received the signal from the gas sensors. The monitoring system was implemented by using CPLD chip to reduce the development time and implement simple one chip system. The time division multi-channel system received the input signal from individual gas sensor at one of divided times by multiplexor and signal processed sequentially. The system reduced the size of peripheral circuit resulted in implementation of creditable simple system.

• Journal of Digital Convergence
• /
• v.12 no.2
• /
• pp.39-43
• /
• 2014

Electronic monitoring system was recently introduced as an effective means of controlling certain criminal offenders. Korean society manages ex-convicts only in reliance on the electronic monitoring system, which has a clear limit. For this reason, The reorientation program should be put before anything to subdivide the perpetrators equipped with the electronic monitoring device and separate them from the others from the start, or identify their criminal propensity and heal them.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.57 no.1
• /
• pp.30-35
• /
• 2008

In this paper, the application of power Quality monitoring based on Sampled Value(SV) which is introduced in the IEC61850 International Standard for substation communication are discussed. Firstly, while Merging Unit(MU) is designed as a process level device transmitting sensor data, the practical problems such as time delay compensation and optical fiber communication are encountered. Secondly, the Sampled Value message which is proper to a power quality monitoring system is presented. Because the power quality monitoring system requests less time-critical service compared to protection and control applications, the Sampled Value service message structure is introduced to improve efficiency. Lastly, the power quality monitoring server having various power quality analysis functions is suggested to verify the performance of Merging Unit With the diverse experiments, it is proved that the process bus distributed solution is flexible and economic for the power quality monitoring.

• Journal of the Semiconductor & Display Technology
• /
• v.18 no.4
• /
• pp.129-133
• /
• 2019

In this paper, we developed a system to monitor the storage capacity of suction-type device such as vacuum cleaner or crop harvesters. The monitoring system consists of load cells and a differential pressure sensor which simultaneously monitor the weight and volume of the stock. Since weighing objects stored in storage containers alone cannot fully monitor the level of filling, more accurate monitoring can be achieved by monitoring volume and fusion with weight information. The volume was monitored using a phenomenon in which the flow rate of the inhaled air varies depending on the volume of the object filled in the storage container. In this paper, we developed a system to monitor the storage in three stages: low, medium and high.

• Proceedings of the KIPE Conference
• /
• 2001.12a
• /
• pp.63-66
• /
• 2001

This paper aims at developing remote control system to control and monitor distributed various devices such as photovoltaic Inverter system through internet. TCP/IP (Transmission Control Protocol/Internet Protocol) and photovoltaic inverter system operated in a row are adopted for network management protocol and applied device, respectively. For controlling and monitoring distributed devices in real-time, java-environment software is constructed. Also, HelloDevice, general-use interface controller between network device and applied device is proposed. Finally, serial communication such as RS-232C is used between controller and applied device.

• Journal of Power System Engineering
• /
• v.20 no.3
• /
• pp.64-73
• /
• 2016

In this research work, an algorithm development on tracking end-point of aerial work platform with jib profile and bend-linked boom was carried out to find the X, Y and Z direction value using coordinate transformation matrix. This matrix consists of device status value(length and angle) based on camera position axis, which are sent from device controller PLUS+1 by CAN protocol. These values are used to measure the distance and angle from the camera to the end-point. Using these distance and angle value, monitoring system controls FAN/TILT/ZOOM status of camera to get an adequate scene of workplace. This program was written in Java, C# and C for mobile device. These results provide the information to the aerial work device for secure operation.

• Proceedings of the KIEE Conference
• /
• 2002.04a
• /
• pp.219-224
• /
• 2002

This paper aims at developing remote control system to control and monitor distributed various devices such as UPS or photovoltaic inverter system through internet. TCP/IP(Transmission Control Protocol/Internet Protocol) and photovoltaic inverter system operated in a row are adopted for network management protocol and applied device, respectively. For controlling and monitoring distributed devices in real-time, java-environment software is constructed. Also, HelloDevice, general-use interface controller between network device and applied device is proposed. Finally, serial communication such as RS-232C is used between controller and applied device.